Interpretable Feature Learning in Multivariate Big Data Analysis for Network Monitoring
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Interpretable Feature Learning in Multivariate Big Data Analysis for Network Monitoring. / Camacho, Jose; Wasielewska, Katarzyna; Bro, Rasmus; Kotz, David.
In: IEEE Transactions on Network and Service Management, Vol. 21, No. 3, 2024, p. 2926 - 2943.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Interpretable Feature Learning in Multivariate Big Data Analysis for Network Monitoring
AU - Camacho, Jose
AU - Wasielewska, Katarzyna
AU - Bro, Rasmus
AU - Kotz, David
N1 - Publisher Copyright: Authors
PY - 2024
Y1 - 2024
N2 - There is an increasing interest in the development of new data-driven models useful to assess the performance of communication networks. For many applications, like network monitoring and troubleshooting, a data model is of little use if it cannot be interpreted by a human operator. In this paper, we present an extension of the Multivariate Big Data Analysis (MBDA) methodology, a recently proposed interpretable data analysis tool. In this extension, we propose a solution to the automatic derivation of features, a cornerstone step for the application of MBDA when the amount of data is massive. The resulting network monitoring approach allows us to detect and diagnose disparate network anomalies, with a data-analysis workflow that combines the advantages of interpretable and interactive models with the power of parallel processing. We apply the extended MBDA to two case studies: UGR’16, a benchmark flow-based real-traffic dataset for anomaly detection, and Dartmouth’18, the longest and largest Wi-Fi trace known to date.
AB - There is an increasing interest in the development of new data-driven models useful to assess the performance of communication networks. For many applications, like network monitoring and troubleshooting, a data model is of little use if it cannot be interpreted by a human operator. In this paper, we present an extension of the Multivariate Big Data Analysis (MBDA) methodology, a recently proposed interpretable data analysis tool. In this extension, we propose a solution to the automatic derivation of features, a cornerstone step for the application of MBDA when the amount of data is massive. The resulting network monitoring approach allows us to detect and diagnose disparate network anomalies, with a data-analysis workflow that combines the advantages of interpretable and interactive models with the power of parallel processing. We apply the extended MBDA to two case studies: UGR’16, a benchmark flow-based real-traffic dataset for anomaly detection, and Dartmouth’18, the longest and largest Wi-Fi trace known to date.
KW - Analytical models
KW - Anomaly Detection
KW - Big Data
KW - Dartmouth Campus Wi-Fi
KW - Data models
KW - Data visualization
KW - Interpretable Machine Learning
KW - Monitoring
KW - Multivariate Big Data Analysis
KW - Network Monitoring
KW - Principal component analysis
KW - Representation learning
KW - UGR’16
U2 - 10.1109/TNSM.2024.3368501
DO - 10.1109/TNSM.2024.3368501
M3 - Journal article
AN - SCOPUS:85186994445
VL - 21
SP - 2926
EP - 2943
JO - IEEE Transactions on Network and Service Management
JF - IEEE Transactions on Network and Service Management
SN - 1932-4537
IS - 3
ER -
ID: 389672967